gravitas_deficiency@sh.itjust.works
on 14 Jul 2024 15:11
collapse
I expect Intel to come off worse in this exchange. GN and L1 are doing some absurd amounts of research and experimentation on this, from what I’ve heard so far. I’m really looking forward to them disproving a bunch of bullshit from Intel. This might even yield a class-action suit.
conciselyverbose@sh.itjust.works
on 13 Jul 2024 17:21
nextcollapse
the team dives into game telemetry data (from Oodle) that shows Intel CPUs represent 70% of the error logs compared to AMD with just 30%.
Uh, what’s the market share? You can’t really provide the one number without the other.
lol IDK how to link to a user on this app, but look at mox right below me. It’s just presented in a confusing way.
70/30% of the logs, not of the errors. It’s equivalent to what you’re thinking of as market share. (I can’t really blame you for misunderstanding, though; this article is poorly written.)
The proportion of errors is better explained in another article:
In fact, for one particular type of error (decompression, a commonly performed operation in games), there was a total of 1,584 that occurred in the databases Level1Techs sifted through, and an alarming 1,431 of those happened with a 13900K or 14900K. Yes – that’s 90% of those decompression errors hitting just two specific CPUs.
As for other processors, the third most prevalent was an old Intel Core i7 9750H (Coffee Lake laptop CPU) – which had a grand total of 11 instances. All AMD processors in total had just 4 occurrences of decompression errors in these game databases.
In case you were thinking that AMD chips might be really underrepresented here, hence that very low figure, well, they’re not – 30% of the CPUs in the database were from Team Red.
conciselyverbose@sh.itjust.works
on 13 Jul 2024 18:13
nextcollapse
Edited.
I just read their one example as one example, not as relative to the 70/30 split of CPUs used.
They still might very well be over-represented because looking at decompression errors doesn’t isolate the CPU, could also be the disk, or RAM. Or even download though that tends to have independent checksumming. And it might not even be he components it could be cosmic rays, if you run code often enough on enough boxes errors are unavoidable, at least on hardware that isn’t space-grade and/or buried underground.
henfredemars@infosec.pub
on 13 Jul 2024 17:44
nextcollapse
With the i9, Intel is screwing their most profitable customers.
RobotToaster@mander.xyz
on 13 Jul 2024 18:13
nextcollapse
It seems crazy that intel were forced into a full recall by the backlash over the FDIV bug in the 90’s, but now they get away with far worse.
irotsoma@lemmy.world
on 13 Jul 2024 19:25
nextcollapse
Not that crazy. Back then they were a luxury item and competition could easily spring up if they lost control of the market. Now they’re a necessity, competition is scarce, and they have deals with most major manufacturers to keep them in businesses even with faulty products. It’s standard late-capitalism consolidation at work.
They were relatively quick to pull the 1.13GHz Pentium III from the market when it turned out to have stability issues from pushing Coppermine too far.
This feels like a repeat of that, but with Intel sticking their heads in the sand and screwing over customers instead.
Kyrgizion@lemmy.world
on 13 Jul 2024 19:01
nextcollapse
Still rocking my 8086K @5Ghz for over 6 years. Best CPU and most stable I’ve ever had, and I’ve been building pc’s since the 90’s (both AMD & Intel). Not willing to shell out for anything new unless this kind of stuff is verifiably fixed.
kusivittula@sopuli.xyz
on 13 Jul 2024 23:27
collapse
my 9600kf was able to do 5ghz with 1.35v for the first year, and i’ve had to lower the clocks again and again as it gets unstable. it’s now only able to do 4.4ghz. recently upgraded to 7800x3d and i’m afraid to see how long this one lasts as it’s 5ghz stock and gets quite warm even in light use. performance is amazing though, realized my old cpu bottlenecked in many games.
I think I’m going with the same processor for my newest build. Seems like a good one.
kusivittula@sopuli.xyz
on 14 Jul 2024 09:06
collapse
it’s a good one. just dont’t get a gigabyte mobo. their bios is comically bad, and i found out that the same issues have continued for many generations…
peopleproblems@lemmy.world
on 13 Jul 2024 19:56
nextcollapse
So. The good news, is that my next work machine is skipping 13 and 14.
The bad news is my next work machine is an Ultra Core processor, we get to be first adopters too! How glorious (/s)!
Alpha71@lemmy.world
on 13 Jul 2024 23:38
nextcollapse
I must have won the silicon lottery because I’ve had no issues with my 13900K
A_Random_Idiot@lemmy.world
on 13 Jul 2024 23:51
nextcollapse
yet
n3m37h@sh.itjust.works
on 14 Jul 2024 01:07
collapse
Dum dum dddduuuuuummm
toddestan@lemm.ee
on 14 Jul 2024 01:43
nextcollapse
I feel lucky too. I have a 14900k that’s stable. I did have some minor stability issues after I built it, but dialing back the motherboard’s idiotic default settings plus a few BIOS updates cleared that up. With that said, if I had to do over, I’d build an AMD system. One of the big reasons I built Intel is that historically my Intel builds have been much more stable and less problematic than my AMD builds.
gravitas_deficiency@sh.itjust.works
on 14 Jul 2024 15:13
collapse
I’ve been running a 5950X since 2020 with nary a hiccup, fwiw. My system before that was an old 2600K that I was just far to lazy to upgrade for a very long time lol
khorak@lemmy.dbzer0.com
on 14 Jul 2024 16:41
nextcollapse
I used my 5900x for over 3 years without a hiccup (months of uptime) when suddenly an update completely wrecked my idle stable. I think I figured it out, but yeah a happy AMD customer with the occasional “Why isn’t shit stable out of the box”
Now Intel… Hahahaha guys I think you got screwed :/ time to lawyer up.
deranger@sh.itjust.works
on 14 Jul 2024 20:47
collapse
The 2500k/2600k was such a good processor. My favorite Intel build I’ve done had a 2500k I overclocked to 4.8GHz back when bitcoin was $10.
threaded - newest
Considering Gamers Nexus track record it’ll be interesting to see how this works out
Intel’s CPUs Are Failing, ft. Wendell of Level1 Techs
I expect Intel to come off worse in this exchange. GN and L1 are doing some absurd amounts of research and experimentation on this, from what I’ve heard so far. I’m really looking forward to them disproving a bunch of bullshit from Intel. This might even yield a class-action suit.
Uh, what’s the market share? You can’t really provide the one number without the other.lol IDK how to link to a user on this app, but look at mox right below me. It’s just presented in a confusing way.
70/30% of the logs, not of the errors. It’s equivalent to what you’re thinking of as market share. (I can’t really blame you for misunderstanding, though; this article is poorly written.)
The proportion of errors is better explained in another article:
Edited.
I just read their one example as one example, not as relative to the 70/30 split of CPUs used.
They still might very well be over-represented because looking at decompression errors doesn’t isolate the CPU, could also be the disk, or RAM. Or even download though that tends to have independent checksumming. And it might not even be he components it could be cosmic rays, if you run code often enough on enough boxes errors are unavoidable, at least on hardware that isn’t space-grade and/or buried underground.
With the i9, Intel is screwing their most profitable customers.
It seems crazy that intel were forced into a full recall by the backlash over the FDIV bug in the 90’s, but now they get away with far worse.
Not that crazy. Back then they were a luxury item and competition could easily spring up if they lost control of the market. Now they’re a necessity, competition is scarce, and they have deals with most major manufacturers to keep them in businesses even with faulty products. It’s standard late-capitalism consolidation at work.
They were relatively quick to pull the 1.13GHz Pentium III from the market when it turned out to have stability issues from pushing Coppermine too far.
This feels like a repeat of that, but with Intel sticking their heads in the sand and screwing over customers instead.
Still rocking my 8086K @5Ghz for over 6 years. Best CPU and most stable I’ve ever had, and I’ve been building pc’s since the 90’s (both AMD & Intel). Not willing to shell out for anything new unless this kind of stuff is verifiably fixed.
my 9600kf was able to do 5ghz with 1.35v for the first year, and i’ve had to lower the clocks again and again as it gets unstable. it’s now only able to do 4.4ghz. recently upgraded to 7800x3d and i’m afraid to see how long this one lasts as it’s 5ghz stock and gets quite warm even in light use. performance is amazing though, realized my old cpu bottlenecked in many games.
I think I’m going with the same processor for my newest build. Seems like a good one.
it’s a good one. just dont’t get a gigabyte mobo. their bios is comically bad, and i found out that the same issues have continued for many generations…
So. The good news, is that my next work machine is skipping 13 and 14.
The bad news is my next work machine is an Ultra Core processor, we get to be first adopters too! How glorious (/s)!
I must have won the silicon lottery because I’ve had no issues with my 13900K
yet
Dum dum dddduuuuuummm
I feel lucky too. I have a 14900k that’s stable. I did have some minor stability issues after I built it, but dialing back the motherboard’s idiotic default settings plus a few BIOS updates cleared that up. With that said, if I had to do over, I’d build an AMD system. One of the big reasons I built Intel is that historically my Intel builds have been much more stable and less problematic than my AMD builds.
I’ve been running a 5950X since 2020 with nary a hiccup, fwiw. My system before that was an old 2600K that I was just far to lazy to upgrade for a very long time lol
I used my 5900x for over 3 years without a hiccup (months of uptime) when suddenly an update completely wrecked my idle stable. I think I figured it out, but yeah a happy AMD customer with the occasional “Why isn’t shit stable out of the box”
Now Intel… Hahahaha guys I think you got screwed :/ time to lawyer up.
The 2500k/2600k was such a good processor. My favorite Intel build I’ve done had a 2500k I overclocked to 4.8GHz back when bitcoin was $10.
I thought I lose the silicon lottery, my 13900KS can’t be underclocked at all or it will BSOD. Now I suppose I’m lucky it work fine at stock settings.
Buy new and charge back. They defraud us.
Don’t they usually check serial numbers?
Not usually, they don’t want to pay staff. Anyways, they’re criminals.
Always buy from a trusted vendor/store, this is more proof of that.
Early 1st gen ryzens segfaulted under certain workloads (particularly during compilations)
So it's not like AMD is stellar that regard. Everyone has a faulty product occasionally.
But AFAIK AMD didn't deny RMAs